Applications of Monolithic Silica Capillary Columns in Proteomics

Applications of Monolithic Silica Capillary Columns in Proteomics Begona Barroso,*,† Dieter Lubda,‡ and Rainer Bischoff† University of Groningen, Department of Bioanalysis & Toxicology, University Centre for Pharmacy, A. Deusinglaan 1, 9713 AV Groningen, The Netherlands, and Merck KGaA, LSP R&D MDA, Frankfurter Str. 250, D-64271, Darmstadt, Germany Received July 10, 2003

The use and applicability of silica based capillary monolithic reversed-phase columns in proteomic analysis has been evaluated by liquid chromatography-mass spectrometry (LC-MS). Chromatographic performance of the monolithic capillaries was evaluated with a tryptic digest of cytochrome C showing very good resolution and reproducibility in addition to the known advantages of a low pressure drop over a time period of 6 months. Monoliths were subsequently tested for their suitability to separate proteins and peptides from samples typically encountered in proteomic research such as in-gel digested tryptic peptide mixtures or fractions of proteolytically digested human serum. The monolithic capillaries also proved useful in the analysis of phospholipid species in bronchoalveolar lavage fluid. Compared to particle-filled conventional capillary columns, rapid and highly efficient separation of peptides and proteins was achieved using these bimodal pore size distribution columns, and good quality collision induced dissociation (CID) mass spectra were obtained on an ion trap mass spectrometer. These novel monolithic separation media are thus a promising addition to the methodological toolbox of proteomics research. Keywords: proteomics • monoliths • Bronchoalveolar Lavage Fluid • serum • cancer • mass spectrometry • HPLC

Introduction Proteomic research aims at studying all of the proteins expressed from a genome thus trying to get a more detailed insight into their biological function and role, as well as to decipher the meaning of the changes observed under different stimuli. The general strategy in proteomics research comprises three main steps: separation, identification, and data interpretation. Regarding separation two-dimensional gel electrophoresis (2D-GE) introduced by O’Farrell1 has been until now the most used methodology due to its high resolving power.2-4 Upon separation protein bands are usually excised and in-gel digested with proteases. The proteolytic peptides are then extracted from the gel and identified by mass spectrometry (MS).5 But the limitations of 2D-GE concerning very small or very large as well as very hydrophobic proteins together with the difficulties in automation, reproducibility and quantification have made researchers look for alternative techniques. Liquid chromatography (LC) coupled to mass spectrometry has become one of the most widely accepted alternatives because of its higher flexibility (different separation modes are available), speed, and ease of automation.6-11 The often limited amount of sample and/or the low abundance of proteins of interest has become a driver for miniaturization of LC in the form of capillary separation systems. * To whom correspondence should be addressed. E-mail: b.barroso@ farm.rug.nl. † University of Groningen, Department of Bioanalysis & Toxicolgy, University Centre for Pharmacy. ‡ Merck KGaA, LSP R&D MDA. 10.1021/pr0340532 CCC: $25.00

 2003 American Chemical Society

Capillary separations, although delivering much improved sensitivity, especially when combined with mass spectrometry,12 often have the drawback of reduced robustness. This is partially due to the limited stability of packed capillary columns and the risk of clogging. New approaches to capillary HPLC, particularly in the field of proteomics, are therefore needed. Monolithic capillary columns made of polymeric13 or silicabased materials14 promise to overcome some of the limitations mentioned above, namely that of packing stability. Furthermore, due to their higher porosity they have less tendency to get clogged. Usually, microcolumns for LC are fabricated by packing beads with a controlled range of diameters and pore sizes. To obtain a better efficiency, columns have been packed with particles of ever smaller diameters15 bringing about another practical limitation: the increase of the back pressure. To circumvent this problem, alternative chromatographic modalities such as ultrahigh-pressure liquid chromatography (UHPC),16 open tubular chromatography (OTC),17 and capillary electrochromatography (CEC)18 have been investigated. All this has led to the use of particle sizes in the range of 3 to 5 µm as a good compromise between column efficiency and pressure drop. In recent years, the introduction of monolithic columns of either organic or inorganic skeleton matrixes has initiated a new era in separation technology. The potential and advantages of monoliths were already recognized by Knox,19 who suggested more than twenty years ago the preparation of a rigid foam inside of a column. Monolithic support structures are formed by in-situ polymerization or consolidation of particulate packJournal of Proteome Research 2003, 2, 633-642

633

Published on Web 09/18/2003

research articles

Figure 1. Microscopic structure of a silica monolith showing the bimodal pore size distribution.

ings. In contrast to conventional beaded columns, where the void volume between individual particles is impossible to avoid, these materials consist of a single piece continuous rod containing a network of highly interconnected pores forming larger and smaller open flow-through channels which allow the mobile phase to flow easily through the column. Moreover, the enhanced permeability of monolithic vs particulate columns results in a much lower back pressure, allowing the use of higher flow rates.20 Different research groups21-24 have tested monolithic materials for their use in HPLC. Hjerten et al.21 and Frechet et al.22 described the preparation of either polyacrylamide or poly(styrene-co-divinylbenzene) stationary phases in the presence of porogens, leading to monolithic materials with a permanent macroporous structure. Monolithic organic polymers have also been successfully prepared within capillaries and applied for HPLC and capillary electrochromatography (CEC).25 However, the chromatographic use of such polymeric materials may have several disadvantages. Most organic polymers, especially those with low degrees of cross linking, are known to swell or shrink in organic solvents leading to dramatic effects on chromatographic performance of these monolithic columns and frequently also to a lack of mechanical stability. Furthermore, the structure of porous organic polymers very often contains micropores, which negatively affect the efficiency and peak symmetry. Nevertheless, organic polymeric monolithic materials offer excellent biocompatibility, are stable over a wide pH range and can be cleaned, without damage, with caustic mobile phases. Such organic polymeric monolith type columns, housed in convenient cartridge designs, are available commercially and have been used for bioseparations with success.26,27 Porous monolithic inorganic materials have been developed to overcome some of the above-mentioned drawbacks. Nakanishi et al.28 developed a new sol-gel process for the preparation of monolithic silica columns with a bimodal pore structure (i.e., with throughpores and mesopores; Figure 1). The method is based on the hydrolysis and poly-condensation of alkoxysilanes in the presence of water-soluble polymers. Tanaka et al.29 demonstrated that this method allows the preparation of chromatographic columns with high efficiencies and low column back pressures as a result of an independent control of the sizes of the silica skeleton and the throughpores. The mean interparticle pore size is indenpendently coupled to the 634

Journal of Proteome Research • Vol. 2, No. 6, 2003

Barroso et al.

particle diameter. The hydrodynamic behavior of silica-based monoliths (and monoliths in general) can be expressed in terms of equivalent particle (sphere) dimensions for the permeability (dperm) and band broadening (ddisp).30 Due to the unique manipulation of their skeleton morphology, silica-based monoliths are characterized by quite different values for dperm (10 µm) and ddisp (2 µm). This means that these silica monoliths offer the separation efficiency of a packed bed of 2 µm porous spheres, but with only the backpressure that a bed of 10 µm spheres will generate, while keeping the adsorption capacity similar to that of a bed of porous spheres. Moreover, it was demonstrated that the recently developed monolithic-type HPLC columns could be operated at high flow rates while maintaining a high efficiency. In this context, Guiochon and Kele31 investigated the reproducibility of the preparation of the first columns getting reproducibilities close and in some cases better than with particle based columns. Due to the small sample volumes and concentrations, which are usually handled in proteomics analysis, a capillary version of a monolithic column is required. This would allow the use of capillary HPLC in conjunction with electrospray ionization tandem MS in a very efficient way, and without the extremely low flow rates required for the conventional beaded capillary columns of the same diameter. On the basis of the commercial development of monolithic silica materials in larger column diameters,15 new silica monolithic materials for the use in nano-LC were developed based on fused silica capillaries of an internal diameter around 100 µm. Such silica-based capillary columns have been successfully used for highly efficient separations of neutral and charged analytes under isocratic elution conditions (e.g., in µLC and CEC).32 Capillary liquid chromatography based on silica monolithic stationary phases was used to screen complex peptide libraries by fast gradient elution coupled on-line to electrospray ionization Fourier transform ion cyclotron resonance mass spectrometry (ESI-FTICR-MS). A slightly modified commercial electrospray interface consisting of a fused-silica transfer capillary and low dead volume stainless steel union at which the electrospray voltage was grounded enabled the effluent of all the capillary columns to be directly sprayed into the mass spectrometer.33 Due to the low back pressure, column length is no longer a limiting factor as with packed columns, allowing to increase separation efficiency (number of plates) and improving the chromatographic result. To this end, it is possible to use monolithic silica capillaries up to 1 m and longer. Because separation efficiency is undoubtedly a critical factor when dealing with highly complex mixtures, we wished to assess the performance of silica-based monolithic capillaries for the analysis of various types of samples currently encountered in proteomics. The presented work shows that this type of columns gives excellent results in terms of separation efficiency, reproducibility and robustness in proteomic applications.

Materials and Reagents Human lung elastin and human neutrophil elastase were purchased from Calbiochem (La Jolla, CA). Horse heart cytochrome C was from Sigma (Zwijndrecht, The Netherlands). Trypsin (sequencing grade, Cat. No. V5111) was obtained from Promega (Promega Benelux B. V., Leiden, The Netherlands).

research articles

Monolithic Silica Capillary Columns

PD-10 columns were purchased from Amersham Biosciences (Uppsala, Sweden). Aurum Serum Protein column was obtained from Bio-Rad (Hercules, CA). Ammonium bicarbonate, EDTA, TFA, sodium acetate, calcium chloride, formic acid, and acetonitrile were from Merck KGaA (Darmstadt, Germany). Serum samples were obtained from cervical cancer patients from the Groningen University Hospital. Bronchoalveolar lavage fluid (BALF) was obtained from COPD patients at the Pulmonology department of the Groningen University Hospital. Preparation of the Different Samples. Cytochrome C Digest. A stock solution of horse heart cytochrome-C (10 mg/ mL) was prepared in ultrapure water and further diluted with aqueous calcium chloride and ammonium bicarbonate to give a final concentration of 1 mg/mL protein in 5 mM CaCl2 and 100 mM ammonium bicarbonate. Then trypsin was added in a ratio 1:40 wt/wt enzyme to substrate and the mixture incubated for 12 h at 37 °C. The resulting tryptic digest was diluted to the required concentration with ultrapure water and injected in the HPLCMS system. Elastin Digest. Fifty microliters of a suspension of insoluble elastin in PBS were mixed with a solution of netrophil elastase in 20 mM ammonium bicarbonate at a protein/enzyme ratio 1000/1 (w/w). The mixture was incubated for 18 h at 37 °C and afterward centrifuged. Two microliters of the supernatant were directly injected in the HPLC-MS system. In Gel Digested Proteins. A protein mixture was separated by polyacrylamide gel electrophoresis and the separated proteins were stained with Coomassie Blue. After destaining the protein band of interest was excised from the gel, washed several times with 25 mM ammonium bicarbonate and acetonitrile, reduced with dithiothreitol, alkylated with iodoacetamide and digested overnight at 37 °C with trypsine (sequencing grade). After digestion the peptides were extracted with a mixture of acetonitrile and formic acid, and the preconcentrated extract was injected in the HPLC-MS system. Serum Depletion and Digestion. Crude serum from a cervical cancer patient (concentration of SCCA1 ) 160.5 µg/L by ELISA) was filtered through a 0.2 µm filter. After filtration, 60 µL were mixed with 140 µL of 20 mM NaH2PO4 pH 7.0 and loaded into the Bio-Rad Aurum Serum Protein column (albumin- and IgG-binding column). The column was vortexed, incubated at room temperature for 15 min and centrifuged. 200 µL of the same buffer were applied on the column and the eluate was collected by centrifugation (depleted serum). All depleted serum (250 µl) was applied on a 1 mL PD-10 column. Next, 150 µl of 25 mM ammonium bicarbonate pH 7.8 were loaded, and the eluent was discarded. The final sample was eluted in 400 µL of the same buffer and collected for tryptic digestion. Serum was digested with sequencing grade modified trypsin (1:100 wt/wt ratio enzyme to substrate) at 37 °C overnight. The digest was prefractionated by SCX using a Poly SEA 2.0 × 150 mm, 5µm, 300 A, column (Michrom BioResources, Auburn, CA) operated at 0.3 mL/min flow rate. The mobile phase comprised two buffers: A: 5mM KH2PO4/H3PO4 pH 3, 25% acetonitrile and B: 5mM KH2PO4/H3PO4 pH 3, 25% acetonitrile, 1.0 M KCl. The salt gradient varied from 0 to 1000 mM KCl (10mM/min). The fraction eluting at the intervale 50-100 mM KCl was collected for further RP-HPLC-MS analysis. Bronchoalveolar Lavage Fluid Preconcentration. 1 mL of BALF supernatant was acidified to pH 3.0 with 10% TFA and

Figure 2. Reproducibility of peptide separations on capillary silica monoliths. Chromatograms from 7 injections performed on a 100 mm i.d. Chromolith CapRod RP18e monolithic silica capillary 15 cm length. 12.5 ng Cytochrome C tryptic digest was separated at a flow rate of 3 L/min. Mobile phase composition: A ) 0.1% formic acid in 2% acetonitrile and B ) 0.08% formic acid in 80% acetonitrile. Gradient: increase 1%B/min from 2% to 40%. Detection: UV-vis λ ) 214 nm.

loaded into a RAM cartridge (LiChrospher RP-18 ADS, 25 µm, Merck KGaA, Darmstadt, Germany) previously conditioned with 5 mL of 0.1% formic acid in ultrapure water. After loading, the column was washed with 2 mL of the same buffer (0.1% formic acid in ultrapure water), and the retained components were eluted with 3 mL of 0.1% formic acid in 60/40 v/v acetonitrile/ water. The eluate was collected and placed in a vacuum centrifuge to eliminate the organic modifier and preconcentrate the sample. A final volume of 50 µL was obtained and 2 µL injected in the monolithic column. Instrumental. The HPLC part of the analytical system consisted of an Agilent Series 1100 capillary LC system (Waldbronn, Germany) comprising a degaser, a binary pump, a thermostated autosampler, and a thermostated column compartment. Chromatographic separation of the peptides took place in a reversephase C18 endcapped monolithic column Chromolith CapRod from Merck KGaA (Darmstadt, Germany). Two different column sizes were used throughout this work: 0.1 mm i.d. × 150 mm length and 0.1 mm i.d. × 500 mm length. Mobile phase A consisted of 0.1% formic acid in ultrapure water. Mobile phase B was 0.1% formic acid in acetonitrile. Flow rates employed varied from 1.5 to 4.5 µL/min. The analytes were detected by a DAD detector at 214 and 280 nm and subsequently with an Agilent SL ion-trap mass spectrometer equipped with an ESI source operated in the positive mode. MS data were acquired over a scan range of 50-1500 amu and 5500 m/z per sec scan rate. Protein identification based on peptide sequence tags was accomplished with software available at the ExPASy Molecular Biology Server of the Swiss Institute of Bioinformatics (http:// www.expasy.org).

Results and Discussion Characterization of the Monolithic Column using Tryptic Peptides. To evaluate the performance and robustness of the newly developed capillary monoliths, a tryptic digest of cytochrome C was analyzed 100 times over a time period of 100 hours (Figure 2). The results show good reproducibility of retention times and peak heights at a back pressure that is considerably lower than for packed columns of the same dimensions. Increasing the column length and thus the overall Journal of Proteome Research • Vol. 2, No. 6, 2003 635

research articles

Barroso et al.

Figure 3. Influence of the length of the column on the separation (LC-MS; Total Ion Current (TIC) over a range of 50-1500 amu): Injected in both cases 2 µL of a tryptic digest of Cytochrome C (0.01 µg/µL). Gradient 1% increase of B (0.1% formic acid in acetonitrile) per min. A ) 0.1 mm i.d. × 150 mm length and B ) 0.1 mm i.d. × 500 mm length Chromolith CapRod RP18e monolithic silica capillary column. Identity of the peaks in Table 1. Table 1 peak no.

identity

1 2 3 4 5 6 7 8 9 10 11 12 13

acetylated GDVEK YIPGTK IFVQK GITWK KTGQAPGFTYTDANK TGQAPGFTYTDANK MIFAGIK EDLIAY TGPNLHGLFGR EDLIAYLK EETLMEYLENPKK EETLMEYLENPK IFVQKCAQCHTVEK

efficiency in terms of plate numbers is an option that is not easily available with packed columns due to back pressure limitations. Highly complex samples such as biofluids from humans often require extremely efficient separation systems to reduce coelution of multiple components. Although modern mass spectrometers may be able to deal with a small number of components simultaneously (e.g., automatic precursor ion selection of 5 ions), ion suppression effects may interfere with obtaining quantitative data, which are very relevant in comparative proteomics studies.34 Figure 3 shows the gain in resolution when going from a 15 to a 50 cm capillary as well as the increased retention of earlier eluting peptides (see Table 1 for peak number identities). Although coelution occurs in some cases on the longer column, the overall quality of the chromatogram is clearly improved. Decreasing separation time is one of the hallmarks of high-efficiency methods. In particular, for complex samples separation time is often a limiting factor especially for LC-MS, where analyses are done sequentially. Due to limitations with back pressure and mass transfer packed 636


capillary columns do not allow to increase the flow-rate significantly without compromising performance and/or column lifetime. The separation performance was checked under different flow rates from 1.5 to 4.5 µL/min. (see Figure 4) corresponding to 19.1 to 57.2 cm/min linear flow velocity for a 100 µm capillary. Due to the favorable flow and mass transfer properties of the monolith, there was no significant effect on the separation, except that the separation time was diminished considerably. The possibility of being able to vary the flowrate over a large area up to very high linear flow velocities combined with the robustness of the monoliths also reduce considerably the “down times” during washing and reequilibration of the column and make these capillaries versatile tools for protein and peptide analysis. The sample loadability in terms of total protein amount retained by the column was calculated by injecting increasing amounts of protein until it appears in the flow-through. For a 0.1 mm i.d. × 150 mm length column 5 µg was the maximum loading capacity. Biological Applications. Separation of Elastin Peptides. Elastin is a quite hydrophobic protein, which gives elasticity to tissues and organs. Due to a high degree of internal crosslinks, mature elastin is a rather stable protein, which is resistant to proteolytic breakdown when compared to most other proteins. However, elastin degradation is known to occur in vivo in a number of diseases that have an inflammatory component.35 It is thus of great interest to study the elastolytic process in more detail in light of developing inhibitors. A number of proteases from various families, named in general elastases, are capable of degrading elastin.36 One of them is human neutrophil elastase (HNE), belonging to the serine protease family. HNE has been studied for many years in conjunction with the development of pulmonary emphy-


research articles

Figure 4. TIC chromatograms corresponding to the injection of 2 µL of Cytochrome C tryptic digest (10 ng/µL) in a Chromolith CapRod RP18e monolithic silica capillary column (0.1 mm i.d. × 15 cm length) run at two different flow rates: A ) 4.5 µL/min and B ) 2 µL/min (the gradient was kept constant).

sema, a chronically developing disease of the airways.37 As part of our studies into Chronic Obstructive Pulmonary Disease (COPD) and the role of elastases in tissue destruction, we are studying the in vitro digestion of human elastin. The resulting degradation products are a complex mixture of peptides covering a very broad molecular weight range: from a few hundred to several thousand Da. This complex sample was used to evaluate the monolithic capillary columns. Figure 5 shows that a range of cleavage products can be separated. On-line MS and MS/MS analysis allowed to identify part of these peptide fragments, which is difficult due to the presence of unnatural amino acids due to the cross-links. Commonly proteins are digested with trypsin for their identification. In this case, the cleavage site can be predicted and the resulting tryptic peptides have basic amino acids which facilitate MS/MS fragmentation and therefore identification. But elastin is composed mainly by hydrophobic amino acids, and when it is digested with neutrophil elastase, the cleavage site is not clearly predictable. The MS/MS spectrum of these peptides shows mainly b ions (Figure 5C). Successful identification is difficult due to the repetitive nature of elastin and requires good chromatographic resolution as that provided by the monolithic column. In the future, the presece of these peptides in different biological fluids will be study to use them as biomarkers for pulmonary diseases. Separation of In-Gel Digested Protein. Typical in proteomics analysis is the identification of proteins separated by gel electrophoresis by mass spectrometry following tryptic digestion. To check the applicability of monolithic columns for this type of samples, a protein mixture was separated by polyacrylamide gel electrophoresis and digested with trypsin. Peptides

in this sample were separated and partially sequenced by MS/ MS in an ion trap mass spectrometer (see Figure 6). Due to the high resolution of the monolithic capillary column, it was possible to select a range of peptides for MS/MS by CollisonInduced Dissociation (CID). As an example, the MS/MS spectrum of the chromatographic peak at 57 min is presented in Figure 6B showing an almost complete series of y and b ions, allowing the sequencing of the peptide and subsequent protein identification. Analysis of Serum from a Cervical Cancer Patient. Discovery of protein biomarkers is another important field in proteomics research. Separation of complex mixtures containing thousands of peptides and proteins such as serum is a challenge for any chromatographic system. A key strategy consists of systematically and selectively reducing the complexity of the sample by, for example, removing high-abundance proteins and by fractionating it into a number of less complex fractions that can be analyzed by LC-MS. A popular combination is strong cation exchange (SCX) with reversed phase chromatography, as these methods are fairly orthogonal.38 The performance of the monolithic capillary column was tested with a subfraction of digested human serum from a cervical cancer patient after cation-exchange HPLC. A major difficulty in analyzing serum is the broad dynamic range of the concentrations of the proteins present. Removal of highly abundant proteins is mandatory to allow the detection of those present in small amounts. As part of our biomarker work, we have compared various ways of depleting serum of albumin and the IgGs.39 Crude serum from a cervical cancer patient was first depleted of albumin and IgGs using a combined dye-ligand/Protein A affinity column and digested with trypsin. Journal of Proteome Research • Vol. 2, No. 6, 2003 637

research articles

Barroso et al.

Figure 5. Human insoluble elastin was digested with human neutrophil elastase. 2 L of the sample were injected in the Chromolith CapRod RP18e monolithic silica capillary column 0.1 × 150 mm. Flow rate 2 L/min. Gradient 1% increase of B/min. (B ) 0.1% formic acid in acetonitrile). A shows the base peak chromatogram (50-1500) and B the extracted ion chromatogram correponding to one of the peptides (m/z 499.2, singly charged) with its MS/MS spectrum displayed in C.

The digest was prefractionated by SCX and submitted to LCMS analysis (Figure 7). As a target protein we selected Squamous Cell Carcinoma Antigen 1 (SCCA1), a marker of squamous cell cervical carcinoma that is used for follow-up on therapy.40 In Figure 7, the resulting base peak chromatogram is presented, together with two extracted ion chromatograms corresponding to expected tryptic peptides of SCCA1. It should also be noted that even with the high salt concentration present in the digest, peak shape and resolution are good. Analysis of Bronchoalveolar Lavage Fluid. Bronchoalveolar lavage fluid (BALF) is collected during fiberoptic bronchoscopy to check the status of the cells and soluble components from 638


the human lung (phospholipids, nucleic acids, and proteins) originated from the thin layer of epithelial lining fluid that covers the airways.41 Analysis of BALF components enables diagnosis and follow-up of lung diseases; thus, there is a great interest in developing methods for this purpose. Two-dimensional gel electrophoresis is the most widely used methodology for the separation of BALF proteins, whereas other techniques based on liquid chromatography are better suited for the analysis of small peptides and low-molecular weight components. BALF is approximately 1000 times less concentrated than serum and contains a high salt concentration due to the washing procedure; therefore, sample pretreatment is required


research articles

Figure 6. (A) TIC corresponding to the tryptic peptides extracted from an in-gel digested protein after separation by SDS-PAGE (Coomassie staining). (B) MS/MS spectrum of m/z 579.1 (doubly charged precursor ion) corresponding to the chromatographic peak eluting at 57 min. in A.

before analysis of BALF. If components below a certain molecular weight range are of interest, there is an easy strategy to eliminate high-molecular weight proteins by restricted access media (RAM) chromatography. One special kind of RAM material is a support belonging to the group of internal surface reversed phase (ISRP) materials, namely the alkyl diol silica (ADS), which was developed a few years ago.42 The principle of the separation mechanism is the access of low molecular mass compounds to the internal surface of the silica particles

(25 µm), on which either butyryl (RP-4 ADS), capryloyl (RP-8 ADS) or stearoyl (RP-18 ADS) moieties are bonded. The external surface of the particles is covered with electroneutral diol groups to prevent adsorption and denaturation of high molecular weight components such as albumin in the sample. The restriction of access is obtained by a so-called physical diffusion barrier, namely the use of silica particles with an appropriate pore diameter of about 6 nm, yielding a molecular mass cutoff around 15 kDa.43 Journal of Proteome Research • Vol. 2, No. 6, 2003 639

research articles

Barroso et al.

Figure 7. Serum from a cervical cancer patient was analyzed after sample pretreatment to remove albumin and IgGs36 and prefractionation by cation-exchange HPLC. Chromatograms (A) Base Peak Chromatogram (BPC); (B) and (C) Extracted Ion Chromatograms of m/z ) 383.2 and 467.1 corresponding to expected peptide fragments of trypsin digested SCCA1 a tumor marker for squamous cell carcinoma. Fraction no. 4 of the cation-exchange run (50-100 mM KCl) was analyzed.

Figure 8. Reversed phase separation on a Chromolith CapRod RP18e monolithic silica capillary column of the low-molecular weight components of a BronchoAlveolar Lavage Fluid (BALF) sample previously preconcentrated by RAM chromatography (A). Marked with arrows are the peaks corresponding to lyso-phosphatidylcholine C16:0 whose extracted ion chromatogram (m/z 496.3) is presented in B. The MS spectrum shows a single dominant ion corresponding to the singly charged protonated molecule (C).

To test the monolithic column, a BALF supernatant was treated with a RAM column and the eluent was injected in the monolithic capillary column. 640


Figure 8 shows that many components have been enriched from BALF by RAM chromatography and can be separated on the capillary monolith. Some of these components have now

research articles


been identified as phospholipid derivatives showing that monolithic reversed-phase capillary columns are also suitable for this compound class. Injection of raw BALF without further pretreatment did not show any peak due to the low concentrations.

Discussion Monolithic columns are a recent addition to the toolbox of the separation scientist. Main advantages are that mass transfer in monoliths is mainly controlled by convection instead of diffusion through the column. Mass transport to the surface inside the skeleton is dominated by diffusion, but this is a very short path length corresponding to that of a 2 µm sphere. As a result, the resolution is almost not affected by the flow rate and high resolution can be achieved at both low and high flow velocities. Silica monolithic capillary columns show excellent separation performance for peptides and proteins, similar to columns packed with 2 µm spherical particles. Peak widths at half the height are on the order of 0.2 min, and the elution of low abundant compounds is not deteriorated in the presence of overloaded compounds, as observed during the chromatographic separation of BALF samples where some components were in a much higher concentration (data not shown). Due to the possibility of using high flow rates, these columns have distinct advantages over packed columns for the separation of very complex mixtures in multidimensional chromatography systems because they allow to shorten the often extensive run times. Although monolithic capillary columns are able to generate a high peak capacity in a short period of time, this usually cannot be fully exploited for the analysis of complex mixtures because peaks are not uniformly spread across the gradient. High complexity samples require the use of shallower gradients producing an increase in the analysis time. In the present work, we have evaluated the performance and suitability of newly developed monolithic silica capillary columns for various proteomics applications. According to our results, these columns present some advantages over packed columns: they are easy to use, and quite robust. After more than 5 months in use and more than 100 runs, there was no evidence of loss of performance. They can be operated at high flow-rates allowing the use of the standard nebulizer instead of nano-interfaces, which are fragile and prone to blocking. These columns are flexible and they show a good performance at both low (1.5 µL/min) and high (4.5 µL/min) flow rates. Because of their capacity to perform fast separations they can be used for fast screening methods and applications in multidimensional chromatography systems. Conditioning and regeneration of these monolithic columns can be done in a short time when compared with the corresponding capillary packed columns, thus making more effective use of costly LC-MS equipment. They can be easily integrated in fully automated systems to perform unattended runs. It is conceivable that many HPLC proteomics applications, running now on conventional particulate columns, could be replaced by monolithic columns.

Acknowledgment. Authors thank Robert Freije for the preparation of the in-gel digested protein sample and Natalia Govorukhina for the preparation of the serum from a cervical

cancer patient. We would like to thank colleagues at the Department of Pulmunology of the Academic Hospital in Groningen for the donation of the BALF samples.

References (1) O’Farrell, P. H. J. Biol. Chem. 1975, 250, 4007-4021. (2) Van Den Heuvel, L. P.; Farhoud, M. H.; Wevers, R. A.; Van Engelen, B. G.; Smeitink, J. A. Ann. Clin. Biochem. 2003, 40, 9-15. (3) Yarmush, M. L.; Jayaraman, A. Annu. Rev. Biomed. Eng. 2002, 4, 349-73. (4) Ong, S. E.; Pandey, A. Biomol. Eng. 2001, 18, 195-205. (5) Godovac-Zimmermann, J.; Brown, L. R. Mass Spectrom. Rev. 2001, 20, 1-57. (6) Wu, S. L.; Amato H.; Biringer, R.; Choudhary, G.; Shieh, P.; Hancock, W. S. J. Proteome Res. 2002, 1, 459-465. (7) Palmblad, M.; Ramstrom, M.; Markides, K. E.; Hakansson, P.; Bergquist, J. Anal. Chem. 2002, 74, 5826-5830. (8) Link, A. J.; Eng, J.; Schieltz, D. M.; Carmack, E.; Mize, G. J.; Morris, D. R.; Garvik, B. M.; Yates, J. R. Nature Biotechnol. 1999, 17, 676. (9) Geng, M. H.; Ji, J. Y.; Regnier, F. E. J. Chromatogr. A 2000, 870, 295-313. (10) Yates, J. R.; Carmack, E.; Hays, L.; Link, A. J.; Eng, J. K. Methods Mol. Biol. 1999, 112, 553-569. (11) Ducret, A.; VanOostveen, I.; Eng, J. K.; Yates, J. R.; Aebersold, R. Protein Sci. 1998, 7, 706-19.(12) Banks, F. J. J. Chromatogr. A 1996, 743, 99-104. (12) Strittmatter, E. F.; Ferguson, P. L.; Tang, K.; Smith, R. D. J. Am. Soc. Mass Spectrom. 2003, 14, 980-991. (13) Chirica G. S.; Remcho V. T. J. Chromatogr. A, 2001, 924, 223232. (14) van Nederkassel, A. M.; Aerts, A.; Dierick, A.; Massart, D. L.; Vander Heyden, Y. J. Pharm. Biomed. Anal. 2003, 32, 233-249. (15) Ishizuka, N.; Minakuchi, H.; Nakanishi, K.; Soga, N.; Nagayama, H.; Hosoya, K.; Tanaka, N. Anal. Chem. 2000, 72, 1275-1280. (16) Wu, N.; Lippert, J. A.; Lee, M. L. J. Chromatogr. A 2001, 911, 1-12. (17) Malik, A. Electrophoresis 2002, 23, 3973-3992. (18) Vanhoenacker, G.; Van den Bosch, T.; Rozing, G.; Sandra, P. Electrophoresis 2001, 22, 4064-4103. (19) Bristow, P. A.; Knox, J. H.; Chromatographia 1977, 10, 279-288. (20) Vervoort, N.; Gzil, P.; Baron, G. V.; Desmet, G. A. Anal. Chem. 2003, 75, 843-850. (21) Zeug, C. M.;. Liao, J. L.; Nakazato, K.; Hjerten, S. J. Chromatogr. A 1997, 753, 227-234. (22) Xie, S.; Svec, F.; Frechet, J. M. J. J. Chromatogr. A 1997, 775, 6572. (23) Minakuchi, H.; Nakanishi, K.; Soga, N.; Ishizuka, N.; Tanaka, N. Anal. Chem. 1996, 68, 3498-3501. (24) Lubda, D.; Cabrera, K.; Minakuchi, H.; Nakanishi, K. J. Sol-Gel Sci. Technol. 2002, 23, 185-189. (25) La¨mmerhofer, M.; Svec, F.; Frechet, J. M. J.; Lindner, W. TRAC 2000, 19, 676-679. (26) Xie, S.; Allington, R. W.; Svec, F.; Frechet, J. M.; J. Chromatogr. A. 1999, 865, 169-174. (27) Premstaller, A.; Oberacher, H.; Walcher, W.; Timperio, A. M.; Zolla, L.; Chervet, J. P.; Cavusoglu, N.; Van Dorsselaer, A.; Huber, C. G. Anal. Chem. 2001, 73, 2390-2396. (28) Nakanishi, K.; Soga, N. J. Non-Cryst. Solids 1992, 139, 1-13, 1424. (29) Minakuchi, H.; Nakanishi, K.; Soga, N.; Ishizuka, N.; Tanaka, N. Anal. Chem. 1996, 68, 3498-3501. (30) Leinweber, F. C.; Lubda, D.; Cabrera, K.; Tallarek, U. Anal. Chem. 2002, 74, 2470-2477. (31) Kele, M.; Guiochon, G. J. Chromatogr. A 2002, 960, 19-49. (32) Tanaka, N.; Nagayama, H.; Kobayashi, H.; Ikegami, T.; Hosoya, K.; Ishizuka, N.; Minakuchi, H.; Nakanishi, K.; Cabrera, K.; Lubda, D. J. High Resolut. Chromatogr. 2000, 23, 111-116. (33) Leinweber, F. C.; Schmid, D. G.; Lubda, D.; Wiesmueller, K. H.; Jung, G.; Tallarek U. Rapid Commun. Mass Spectrom. 2003, 17, 1180-1188. (34) Annesley, T. M. Clin. Chem. 2003, 49, 1041-1044 (35) Colburn, K. K.; Langga-Shariffi, E.; Kelly, G. T.; Malto, M. C.; Sandberg, L. B.; Baydanoff, S.; Green, L. M. J. Investig. Med. 2003, 51, 104-109. (36) Mecham, R. P.; Broekelmann, T. J.; Fliszar, C. J.; Shapiro, S. D.; Welgus, H. G.; Senior, R. M. J. Biol. Chem. 1997, 272, 18 07118 076. (37) Shapiro, S. D. Biochem. Soc. Trans. 2002, 30, 98-102. (38) Adkins, J. N.; Varnum, S. M.; Auberry, K. J.; Moore, R. J.; Angell, N. H.; Smith, R. D.; Springer, D. L.; Pounds, J. G. Mol. Cell Proteomics 2002, 1, 947-55.

Journal of Proteome Research • Vol. 2, No. 6, 2003 641

research articles (39) Govorukhina, N.; Keizer-Gunnink, I.; van der Zee, A. G. J.; de Jong, S.; de Bruijn, H. W. A.; Bischoff, R. J. Chromatogr. A, in press. (40) Esajas, M. D.; Duk, J. M.; de Bruin, H. W.; Aalders, J. G.; Willemse, P. H.; Sluiter, W.; Pras, B.; ten Hoor, K.; Hollema, H.; van der Zee, A. G. J. J. Clin. Oncol. 2001, 19, 3960-6. (41) Noe¨l-Georis, I.; Bernard, A.; Falmagne P. Wattiez, R. J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 2002, 771, 221-236.

642


Barroso et al. (42) Boos, K. S.; Rudolphi, A.; Vielhauer, S.;.Walfort, A.; Lubda, D.; Eisenbeiss, F. Fresenius J. Anal. Chem. 1995, 352, 684-690. (43) Racaityte, K.; Lutz, E. S. M.; Unger, K. K.; Lubda, D.; Boos, K. S. J. Chromatogr. A 2000, 890, 135-144.

PR0340532

Applications of Monolithic Silica Capillary Columns in Proteomics

Recommend Documents